Abstract
The Gaussian processes (GP) model has been successfully applied to the prediction of nonstationary time series. Due to the model’s covariance function containing an undetermined hyperparameters, to find its maximum likelihood values one usually suffers from either susceptibility to initial conditions or large computational cost. To overcome the pitfalls mentioned above, at the same time to acquire better prediction performance, a novel multi-scale Gaussian processes (MGP) model is proposed in this paper. In the MGP model, the covariance function is constructed by a scaling function with its different dilations and translations, ensuring that the optimal value of the hyperparameter is easy to determine. Although some more time is spent on the calculation of covariance function, MGP takes much less time to determine hyperparameter. Therefore, the total training time of MGP is competitive to GP. Experiments demonstrate the prediction performance of MGP is better than GP. Moreover, the experiments also show that the performance of MGP and support vector machine (SVM) is comparable. They give better performance compared to the radial basis function (RBF) networks.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Sollich, P., Halees, A.: Learning curves for Gaussian process regression: approximations and bounds. Neural Computation 14, 1393–1428 (2002)
Mackay, D.J.C.: Introduction to Gaussian processes. In: Gerstner, W., Hasler, M., Germond, A., Nicoud, J.-D. (eds.) ICANN 1997. LNCS, vol. 1327. Springer, Heidelberg (1997)
Sollich, P.: Bayesian methods for support vector machines: Evidence and predictive class probabilities. Machine learning 46, 21–52 (2002)
Bellhouari, S.B., Bermak, A.: Gaussian process for nonstationary time series prediction. Computational Statistics and Data Analysis 47, 705–712 (2004)
Basseville, M., Basseville, A., Chou, K.C.: Modeling and estimation of multiresolution stochastic processes. IEEE Trans. on Information Theory 38, 766–784 (1992)
Fitzek, F.H.P., Reisslein, M.: MPEG4 and H.263 video traces for network performance evaluation (extended version). Technical Report: TKN-00-06. TU Berlin Dept. Electrical Engineering, Telecommunication networks Groups (2000)
Moody, J., Darken, C.: Fast learning in networks of locally-tuned processing units. Neural Computation 1, 281–294 (1989)
Vapnik, V., Golowich, S., Smola, A.J.: Support vector method for function approximation, regression estimation, and signal processing. In: Neural Information Processing System (NIPS), pp. 322–327. MIT press, Cambridge (1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhou, Y., Zhang, T., Li, X. (2006). Predicting Nonstationary Time Series with Multi-scale Gaussian Processes Model. In: King, I., Wang, J., Chan, LW., Wang, D. (eds) Neural Information Processing. ICONIP 2006. Lecture Notes in Computer Science, vol 4232. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11893028_57
Download citation
DOI: https://doi.org/10.1007/11893028_57
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-46479-2
Online ISBN: 978-3-540-46480-8
eBook Packages: Computer ScienceComputer Science (R0)